Maria Georgescul, Manny Rayner, P. Bouillon, Nikos Tsourakis
{"title":"Discriminative learning using linguistic features to rescore n-best speech hypotheses","authors":"Maria Georgescul, Manny Rayner, P. Bouillon, Nikos Tsourakis","doi":"10.1109/SLT.2008.4777849","DOIUrl":null,"url":null,"abstract":"We describe how we were able to improve the accuracy of a medium-vocabulary spoken dialog system by rescoring the list of n-best recognition hypotheses using a combination of acoustic, syntactic, semantic and discourse information. The non-acoustic features are extracted from different intermediate processing results produced by the natural language processing module, and automatically filtered. We apply discriminative support vector learning designed for re-ranking, using both word error rate and semantic error rate as ranking target value, and evaluating using five-fold cross-validation; to show robustness of our method, confidence intervals for word and semantic error rates are computed via bootstrap sampling. The reduction in semantic error rate, from 19% to 11%, is statistically significant at 0.01 level.","PeriodicalId":186876,"journal":{"name":"2008 IEEE Spoken Language Technology Workshop","volume":"3 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 IEEE Spoken Language Technology Workshop","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SLT.2008.4777849","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 4
Abstract
We describe how we were able to improve the accuracy of a medium-vocabulary spoken dialog system by rescoring the list of n-best recognition hypotheses using a combination of acoustic, syntactic, semantic and discourse information. The non-acoustic features are extracted from different intermediate processing results produced by the natural language processing module, and automatically filtered. We apply discriminative support vector learning designed for re-ranking, using both word error rate and semantic error rate as ranking target value, and evaluating using five-fold cross-validation; to show robustness of our method, confidence intervals for word and semantic error rates are computed via bootstrap sampling. The reduction in semantic error rate, from 19% to 11%, is statistically significant at 0.01 level.